CHAPTER

FORCES
I don't know what I may seem to the world, but, as to myself, I seem to
have been only a boy playing on the sea shore, and diverting myself in now
and then finding a smoother pebble or a prettier shell than ordinary, whilst the
great ocean of truth lay all undiscovered before me.
Isaac Newton

8.1 THE FUNDAMENTAL FORCES: CLASSIFICATION AND UNIFICATION

Using crude water clocks to time balls rolling down inclined planes, Galileo searched
for and found a description of how bodies fall. His Jaw of falling bodies, however, wasn' t
a fundamental law of nature. Within half a century it was superseded by a deeper insight
into nature - Newton's universal law of gravity. Through the genius of Newton, the
force of gravity,
163
164 FORCES

F = - G Ml~2 r. (8.1)
r

was revealed as a fundamental force of nature.

Inspired by Newton, physicists in the eighteenth century sought to identify, classify,
and mathematically describe the numerous forces observed in nature . Knowledge of these
forces provided physics with a certain predictive power, because according to Newton's
second law, F = ma, forces shape the motion of all things. Through painstaking ex-
periments, these physicists reached empirical descriptions of forces in the world about
them: tensions, spring forces, friction , viscosity, electricity, magnetism, chemical action.
As the number of forces grew, so did the applications in an increasingly industrialized
world. Yet a question which confronted these physicists was: Are all these forces fun-
damental, or can they be reduced to more basic forces?
Not until late in the eighteenth century did a second force emerge as fundamental -
the electric force. The French engineer Charles Augustin Coulomb assumed that, anal-
ogous to the gravitational force between two masses, the electric force between two
charges is proportional to the product of the charges. Experimentally, he found that the
electric force is s imilar to gravity in another way: the force between two charges decreases
as the square of the distance between them. Summarized mathematically, the electric
force F between two charges q 1 and q 2 which are separated by a distance r is known as
Coulomb's law and written as

(8.2)

Just as G is a universal constant for gravity , Ke is a universal constant for electricity.

Magnetism was also identified as a fundamental force of nature. The attractjon or
repulsion between two magnets could be described by a force between pairs of magnetic
poles. The progress of physics appeared to be a triumph of Newtonian mechanics: the
forces of nature were successively reduced to attractions and repulsions between particles.
The first 40 years of the nineteenth century , however, saw a growing reaction against
such a division of phenomena in favor of some kind of correlation of forces. The tum
inward to unification of forces was spearheaded by Oersted , Ampere, and Faraday. By
the middle of the century they had succeeded in unifying two hitherto disparate forces ,
electricity and magnetism, into one - electromagnetism. The process was crowned in the
theory of James Clerk Maxwell , who expressed the unification by a set of equations
which interrelate electric and magnetic phenomena. Soon tensions, spring forces, friction ,
viscosity , chemical actions , and even light were recognized as arising fundamentally from
the electromagnetic force. Based on Maxwe ll 's success the search for a common math-
ematical description, or unification, of forces had begun.
With the twentieth century came the discovery of radioactivity, the probing of atoms,
and the subsequent realization that more than just gravity and electromagnetism would
be needed to explain this new world. Both a new dynamics - the laws of quantum
mechanics, which supersede Newton 's laws on atomic and subatomic distance scales-
8.2 THE STRENGTH OF GRAVITATIONAL AND ELECTRIC FORCES 165

and new forces were needed. Because Newton's laws do not apply on the distance scales
at which the new forces act, we shaJI not pursue the study of these forces in this book.
Nevertheless many themes, such as the quest for the fundamental constituents of matter
and the fundamental forces of nature, carry over from classical mechanics, and it is
interesting to survey briefly what has been found.
Experiments probing atoms revealed that inside an atom there is a compact center -
the nucleus - composed of positively charged protons and neutral neutrons. Negatively
charged electrons orbit the nucleus, held by the electric force from the protons. The
natural question which arose is: What holds the nucleus together? Physicists realized that
neither gravity nor electromagnetism held the compact nucleus together, but that a new
force was at work. Aptly named the strong force, it overcomes the electric repulsion
between protons and holds the nucleus together. Unlike gravity and electricity, the strong
force does not extend to great distances; it has a limited range - the size of a nucleus,
10- 13 em. Outside this range, the strong force has virtually no effect. If it did, we
wouldn't be here; matter would collapse into dense lumps of subatomic particles.
Natural radioactivity could not be explained in all instances by any of the known
forces - strong, electromagnetic, or gravitational. Another force was implicated in the
decay of nuclei - the weak force. This force is intrinsically weaker than the strong force ,
and has an even more limited range, about 10- 16 em or 1/ lOOOth the size of the nucleus.
Because of its severely limited range its most common manifestations in nuclei are feeble
indeed, about 106 times weaker than the strong force. Nevertheless, the weak force plays
an essential role in the release of nuclear energy in stars , and in causing some stars
ultimately to explode.
The behavior of the four fundamental forces of nature - strong, electromagnetic,
weak, and gravitational - is reasonably well understood, but nobody knows why there
should be four of them. Albert Einstein spent the last 20 years of his life unsuccessfully
searching for a way to unify two of the forces, gravity and electromagnetism. Twentieth-
century physics has become a story of attempting to explain all the complexities of physics
as aspects of similar systems, a search for unification of the fundamental forces. The
water clocks and inclined planes of Galileo have been replaced by increasingly larger,
more energetic particle accelerators. Unified theories are emerging that bring together
the weak and electromagnetic forces, as well as more comprehensive theories that attempt
to give a coherent account of how aJI these forces may have evolved from simpler laws
in the infancy of the universe. The early universe ultimately may be the only experimental
test for such theories. It may be the great ocean of truth that still lies undiscovered before
us.

8.2 THE STRENGTH OF GRAVITATIONAL AND ELECTRIC FORCES

One of the great and deep mysteries of physics is that the laws describing the gravitational
force and the electric force have essentially the same mathematical character. They are

F -G M1M2 r (8.1)
r2
and

F K q 1 q2 r (8.2)
e r2
"166 FORCES

It seems almost a minor point that each law contains an unspecified universal constant.
Yet for applications to the real world, it is essential to know what those constants are.
For gravity, we already know that G is related to the acceleration g of a falling body
near the surface of the earth and the mass and radius of the earth:
(7.11)
The radius of the earth has been known since antiquity, and we also know g, so measuring
G immediately tells us the mass of the earth ME. The determination of G was one of the
great classic experiments in physics.
Henry Cavendish, a British physicist, performed the historic measurement of G in
1798. Deeply inspired by Newton, Cavendish regarded the Principia as the model for
exact sciences, and the search for the forces between particles guided his scientific
explorations. But Cavendish had fitful habits of publication; he left unpublished whatever
did not fully satisfy him. Luckily the determination of G was an experiment in which he
took pride.
Figure , which is adapted from Cavendish's 1798 article, shows the apparatus he
invented for his delicate experiment which measured forces equal to one-billionth of the
weights of the bodies involved. The two small lead balls are attached to a rigid rod
forming a dumbbell which is suspended by a thin fiber that allows the dumbbell to rotate
freely. When the two larger lead balls are placed near the ends of the dumbbell, the
smaller masses are attracted to the larger ones by the gravitational force F . This force,
although extremely small , nevertheless exerts a torque which rotates the dumbbell and
twists the fiber. The fiber opposes the twisting with a torque that equals the angle of
rotation times a known coefficient. By measuring the deflection of a beam of light reflected
off the small mirror attached to the fiber, the angle and thus the force F can be determined.
Because the balls are spherical, their masses act as if concentrated at their centers.
Knowing the masses m and M, and the distance r between their centers when the twisting
stops, we can calculate the value of G through Eq. . As stated in Chapter , that
value is found to beG = 6.67 X w- N m 2/kg 2

----~light
--------
------------~

OM
Figure 8.1 Schematic of Cavendish's apparatus for measuring G.

By this experiment, Cavendish rendered the universal law of gravitation complete.

The law was no longer a proportionality relation as Newton had stated it, but an exact
8.2 THE STRENGTH OF GRAVITATIONAL AND ELECTRIC FORCES 167

Jaw through wh ich quantitative analysis could be made. It was the most important con-
tribution to the study of gravitation since Newton.
Through a similar experiment in 1787, Coulomb showed that the electric force
between two charges is analogous to gravity: it decreases as the inverse square of the
distance between the charges. Likewise the value of the electric constant Ke could be
measured; its value is 9.0 x 109 N m2/C 2 , where 1 C, one coulomb, is the unit of
charge. But what is charge?
The early Greeks had discovered that amber attracts bits of straw and identified that
property of amber with charge. Charge is that which creates electric force; even today
we can reduce its essence no further . We don' t know exactly what charge is any more
than we know what mass is. Unlike mass , charge comes in two different varieties -
positive and negative - terms introduced by Benjamin Franklin. Like charges repel
whereas opposite charges attract. On the other hand , gravity is always attractive. There
is also a fundamental unit of charge - the charge of the proton (or, exactly equal in
magnitude and opposite in sign, the charge of the electron). All charges come in mumtiples
of this unit of e lectricity; the charge of a proton is 1.6 x 10- 19 coulombs.
Atoms consist of positively charged nuclei and negatively charged electrons. An
electron is about 2000 times lighter than a proton. The simplest atom, hydrogen, consists
of one proton and one e lectron separated by w-s em. This separation distance is much
larger than the nucleus , so we picture a negative charge outside the positive nucleus . The
force that holds the electron to the proton to make a hydrogen atom is the electric force.
To construct heavier atoms in our model, we first construct nuclei with more protons
and neutrons in them. Since the overall charge of atoms is zero, there are as many
e lectrons around the nucleus as there are protons in the nucleus. Although the electrons
repel each other, they are held to the nucleus by the electric force.
Atoms in tum can attract other atoms to make larger composites called molecules.
The force that holds the atoms together to form molecules is again the electric force, in
this case a residue of it which extends outside the basically neutral atom when its electron
orbits are distorted by the presence of neighboring atoms . Atoms and molecules can form
larger agglomerations which we see as liquids and solids. These too are held together by
electric forces.
Given the s.trengths and nature of the various fundamental forces, we can ascertain
the phenomena rthat each of them controls. Gravity acts on all matter, as reflected by its
dependence on the masses of objects. Although its strength diminishes universally with
distance, the effects of gravity are nevertheless felt across the far reaches of the universe.
Gravity holds together planets and stars, organizes solar systems and galaxies; it orders
the universe.
Electricity, in giving rise to tensions, spring forces, friction, viscosity, and chemical
actions, governs the everyday world around us. It is intrinsically stronger than gravity
and is the dominant force on our size scale, for objects as large as mountains and as
small as atoms. On larger scales, gravity dominates because the gravitational attractions
of all individual masses add whereas the electrical attractions and repulsions of individual
electrons and protons tend to cancel. The cancelation is almost perfect in a large body ,
essentially because of the very strength of the electric force, which causes charge to flow
rather easily and neutralize most charge excesses.
On distance scales of nuclear size or smaller, the strong and weak forces organize
matter. The stro ng force , intrinsically stronger than gravity and electromagnetism, dom-
168 FORCES

inates within this distance. Physicists speculate that the lack of influence of the strong
force outside nuclei is analogous to the relative unimportance of the electrical force on
scales larger than the earth: the strong force is so strong that it neutralizes its sources
within the nucleus essentially perfectly.
Table summarizes the four fundamental forces of nature, their respective ranges,
and their respective strengths (as estimated for the forces acting between two protons at
short range). Note the utter unimportance of gravity on the atomic scale of w-s em.
Only on the scale of a mountain, I km or 10 13 atoms on a side, I km 3 or 1039 atoms in
volume, does the mass become large enough to compensate the relative strength factor
of 10- 39 .

Table 8.1 Characteristics of the Four Fundamental Forces

Force Relative strength Range Importance

Strong w- 13 cm Holds nucleus together

Electromagnetic Lnfmite Controls everyday phenom-

ena - friction, tensions, etc.

Weak w-2 10- 16 em Nuclear transmutation

Gravitational 10 39 i nfinite Organizes large-scale phe-

nomena and universe

8.3 CONTACT FORCES

The fundamental forces of nature are action-at-a-distance forces: their effects can be
experienced when the particles are not in contact. A second category of forces is contact
forces. These are forces which two objects exert on each other when they are physically
in contact with each other, as for example, when a book rests on a table.* Contact forces
are not fundamental forces; instead, they arise from electric forces acting in complicated
ways among enormous numbers of atoms.
The contact forces are described by empirical rules, which are experimental sum-
maries of the net result of all the complications. Most of these empirical descriptions
were deduced by eighteenth-century scientists who were unaware of the underlying fun-
damental forces. Even today, when the fundamental forces are better understood, quan-
titative deduction of the contact forces from fundamenta l laws remains difficult and the
empirical rules are still commonly used.

*The implication that contact forces drop abruptly to zero when two bodies separate, though valid on
macroscopic distance scales, is an idealization. On the atomic scale the contact force falls off continuously
though rapidly when the distance of separation increases, as one would expect from the elecuical origin of the
force.
8.3 CONTACT FORCES 169

The tension in a rod, wire, rope, or string is an example of a contact force. The
atoms in any of these objects are bound together by electric forces which tend to keep
the atoms a certain distance apart, called the equilibrium distance. When you pull on one
end of the wire or rope each atom electricaJly tugs on its neighbors, and the pull is
transmitted to the other end, much as in a chain link. The net result of all the electric
forces acting on the atoms is the macroscopic force- the tension. The foregoing argument
provides the atomic explanation of the empirical rules stated for tension in Chapter :
tension acts along the wire or rope, and if the weight of the object is negligible or has
no component along its length, the pull is transmitted to the other end undiminished in
strength . If the weight cannot be neglected, the tension varies along the wire or rope in
such a way as to hold it up against gravity.
Compression in a rigid rod is another example of contact force. When you push on
one end, each atom pushes on its neighbors, attempting to maintain the equilibrium
separation, and the push is transmitted to the other end of the rod.
In addition to their role in transmitting forces, tension and compression also change
the body sustaining them, elongating or shortening it. We can obtain a qualitative picture
of what is happening by carrying our previous reasoning about electric forces a step
further. Consider, for example, a wire under tension. As stated before, the atoms in the
wire have an equilibrium position at which electric forces tend to keep each atom. The
new point we now wish to e mphasize is that when the wire is put under tension, each
atom is pulled a tiny bit away from the equilibrium distance. The electric forces try to
put the atoms back into the equilibrium position, but some elongation remains while the
wire is under tension.
The empirical law for the change in length of a wire is simple: the force exerted by
a wire (on an object) is proportional to the change in length of the wire. This is known
as Hooke's law (after Robert Hooke). Hooke's Jaw also applies to compressions.
Springs obey Hooke's law especially well. Here the elongation is greatly increased
by a geometrical effect, the straightening out of the coiled wire.
Hooke's Jaw is expressed mathematically as follows:

Hooke's law:
F = - kx, (8.3)

where x is the change in length of the wire or spring from its equilibrium value. ln
discussing Eq. one often uses the spring as a model , and adopts a terminology
associated with springs to describe the stretching of straight wires and many other objects
as well as springs. The constant k is called the spring constant and is a measure of the
stiffness of the s pring; the stiffer the spring, the larger the value of k . The direction of
the force always opposes the displacement of the end of the spring from its unstretched
position. When x > 0, the spring is stretched and F is negative; when x < 0, the spring
is compressed and F is positive. The spring force always acts to restore the spring to its
unstretched length, as Fig. iJiustrates.
170 FORCES

~equilibrium length

Figure 8.2 The force exerted by a spring described by Hooke's law.

In addition to tension and compression, which we already made use of in Chapter

, and Hooke's Jaw relating tension and compression to change of length, there are further
varieties of contact force which will play a role in the applications of Newton's laws in
the present chapter. For example, whenever any object is pressed against another there
is a contact force between the two objects, known as the normalforce. This force is a
result of repulsion between the atoms of the two o bjects. The empirical rules for a normal
force are that its magnitude depends on bow bardl the object is pressed, but the direction
of the normal force acting on an object is always perpendicular to the surface.* Figure
indicates normal force" between different obj ects.
Other important examples of contact forces ilnclude friction and viscosity. We shall
discuss these forces in subsequent sections.

(a) (b) (c)

Figure 8.3 lllustration of the normal force exerted by (a) wall against
hand, (b) hand against tray, (c) floor against person.

*The concept of normal force can also be applied wnen one object is pulled away from another in a
direction normal to the surface and the pull is resisted by adhesion. ln this case the force is the result of
attraction between atoms.
8.3 CONTACT FORCES 171

To bring out in more detail how contact forces arise microscopically, a typical electric
force between two neighboring atoms is plotted in Fig. . One sees immediately that
the simplicity of the fundamental llr 2 force between two charges has gotten submerged
in complications. The attraction at long range is the resultant of canceling attractions and
repulsions among the electrons and nuclei of the two atoms; it falls off with distance as
7
, - or so. The repulsion at short range is governed by quantum effects which tend to
keep the electrons of the two atoms apart.

f(r)

Figure 8.4 The dependence of the force f(r) betweelll two atoms on
their separation distance r, and linear approximationf'(r0 )(r - r0 ) to
f(r) near r0

In the absence of external disturbance the equilibrium distance in Fig. is r0 ,

where the electric force vanishes. Greater separation distances, such as occur when a
wire is pulled from one end, are resisted by the attractive long-range force, giving rise
to tension. Closer spacing, such as occurs under compression, is resisted by the short-
range repulsion giving rise to the normal force.
The microscopic basis for Hooke's law can be found by drawing a straight line in
Fig. tangent to the force curve at r 0 . This linear relation between force and displace-
ment corresponds to Hooke's law, and it is a good approximation to the force near the
equilibrium point.
What factors determine the spring constant k? If we consider tbe stretching of a
straight wire of length L by an attached weight, as pictured in Fig. , the force on one
atom in the wire due to its neighbor is

Figure 8.5 Stretching of a loaded wire.

172 FORCES

f = f'(r0 )(r - r0 )

= -If' (ro)l (r - r o) (8.4)

wheref(r 0 ) is the (negative) slope of the force curve at r0 The total force of tensiion on
all the atoms in a cross section of area A is of order*

F = - ~ lf'(r0 )l(r - r 0 ), (8.5)

Ao
where A0 is the area of cross section occupied by a single atom. The overall stretching
of the wire, IlL , is the stretching between a neighboring pair of atoms, r - r 0 , times
the total number of such stretchings in length L , namely L/L0 , where L0 is the separation
between neighboring atoms. Thus we have IlL = (r - r0 )LIL0 orr - r0 = L() tJ.L/L
and

F = - ~ If' (ro)l Lo IlL

A0 L
or

F = _1/'(ro)ILo A IlL = _ YA IlL . (8.6)

A0 L L
The quantity Y is called Young's modulus and contains the dependence on microscopic
properties [L0 /A 0 = r0 !rJ = r0 1 and j'(r 0 )] of the particular atoms in the wire. The
dependence on geometrical properties has been separated out in the factor A tJ.L/L. Hooke's
constant k for the straight wire is YA!L; note that it depends on both geometrical and
microscopic factors. To produce an ordinary spring, one coils the wire. This makes it
much less stjff (smaller k , larger tJ.L = x for a given pull) because of an extra geometrical
effect: most of tJ.L comes from straightening the coil.
The dependence of k on f'(r0 ) (which depends on the material used) and on the
geometry of the wire is a reminder that the spring force is not fundamental. But not until
a wire is stretched by a large amount does the full complexity of the contact force become
evident. Beyond some point, the force in Fig. is no longer well represented by the
linear approximation in Hooke's law. Moreover, when stretched too far, the wire does
not return to its original length when the applied force is removed; the stretching has
caused permanent deformation by moving or creating imperfections in the wire which
reduce the number of atomic bonds in a cross-sectional area. The nature of such imper-
fections, and the eventual breaking point of the wire, depend not only on the interatomic
force law and the overall geometry of the wire, but also on the history of how the wire
has been treated .

8.4 APPLICATION OF NEWTON'S LAWS

Now that we have increased our repertoire of empirical rules for forces , we return to
their use in Newton's laws. The success of Newtonian mechanics was the identification

*This relatiom is only approximate because each atom is pulled by several other neighboring atoms at
various angles with respect to the direction along the wire.
8.4 APPLICATION OF NEWTON'S LAWS 173

of forces and the subsequent dynamical explanatlion of the motion of objects influenced
by these forces. Generations of students have learned Newton's laws by solving all sorts
of dull as well as interesting problems in meclhanics. Even before Newton the great
sixteenth-century humanist Erasmus, when he was a student, wrote a letter to a friend
saying how dull mechanics lectures were. Nobody will ever know how many minds,
eager to learn the secrets of the universe, found themselves studying inclined planes and
pulleys instead, and decided to switch to some more interesting profession. Nevertheless,
through solving problems, you do understand physics.
Our task is to apply Newton's laws and analyze the motion of objects. Let's list a
few extremely useful steps which, once mastered, will allow you to solve a huge class
of problems in mechanics. These duplicate in part the steps suggested in Section for
treating the special case of equilibrium, but the procedures are so important that they
bear some repetition:

1. Draw a free-body diagram for every object whose motion is to be analyzed.

This entails drawing each object separated from all others and clearly indicating
the external forces acting on each by arrows which start or end on the object.
2. Label all external forces acting on the objects. In labeling, use the fact that
action-reaction pairs of forces are equal and opposite.
3. Choose a coordinate system for each object under consideration. The choice
should simplify the equations as much as possible; for example, it is often useful
to place one axis along the direction of the acceleration.
4. Apply Newton's second Jaw in component fonn to each object:
2: F, = rna, .
This requires resolving forces into components.
5. Write down, or use directly in the second law, any equations of constraint that
restrict the motion of the objects under analysis. For example, the vertical
component of acceleration for an object constrained to slide along a floor van-
ishes. Similarly, if two weights are connected by an extensionless cord running
over a pulley, their accelerations are equal in magnitude.
6. By now you should have as many equations as unknown quantities. Whenever
possible first solve for the unknowns algebraically, then substitute numbers and
units to obtain quantitative answers.

The following examples illustrate the method used to apply Newton's laws. Each
body in these examples is treated as a point mass so that the forces are assumed to act
at one point and there is no torque (we shall deal with some problems involving torque
in later sections). In addition, the masses of pulleys and strings are considered negligible.
Although these assumptions may appear artificial (where can you buy a massless string?),
it is understanding the method which is important now.

Example 1
A 60-kg passenger is riding in an elevator. Find the force exerted by the floor on the
passenger when the elevator is
174 FORCES

(a) accelerating upward at 3.0 m/s 2 ,

(b) accelerating downward at 3.0 m/s 2 , and
(c) moving downward with a constant speed of 4.0 m/s.
In drawing a free-body diagram of the passenger (unflatteringly represented by a
block) we have only two forces to consider, gravity mg and the normal force N.

If we choose the positive z axis to point vertically upward, the second law L F = ma
1
implies
N- mg = ma,

which tells us that the normal force is N = m(g + a). For case (a) we use a = + 3.0
mls 2 and get Na = 770 N. For case (b), the acceleration is downward and therefore
negative, a = - 3.0 m/s 2 . Substituting this into our expression for N, we get Nb =
410 N. For case (c) the acceleration is zero. and the normal force is simply equal to the
weight: Nc = 590 N.

Example 2
A parcel slides down a chute so smooth that friction is negligible. Calculate the acceleration
of the parcel as well as the normal force from the chute.

The free-body diagram for the parcel is shown below. Note that the normal force N
is perpendicular to the inclined chute.
8 .4 APPLICATION OF NEWTON'S LAWS 175

Choosing the x axis to point down the plane, in the direction of the acceleration, we have
the following:
L Fx = ma implies mg sin e= ma,
L F>' = 0 implies N - mg cos e 0.
The second equation tells us that the normal force is N = mg cos e. The first equation
tells us that the acceleration of the parcel is a = g sin e. This is the relation that enabled
Galileo to determine g. Motion on an inclined plane is uniformly accelerated, but can be
made much slower than free fall , and thus more easily measured, by taking e small.

Example 3
Two blocks, one of mass m 1 = 1.0 kg, the other with mass~ = 2.0 kg, are pushed
along a frictionless surface by a force of 2.0 N. Find the acceleration of the blocks and
the force of block 1 on block 2.

Following the steps outlined in the text, we first draw a free-body diagram for each
block. The forces acting on block l are gravity m 1g; normal force N 1 ; outside push F ;
and , because l pushes on 2, 2 pushes back on l (by the third law) with a force we' ll
call P. A similar set of forces act on block 2, as shown in the free-body diagram. Note
that F does not act directly on block 2, so it is not shown acting on it. The effect ofF
is felt indirectly, through the force P arising from contact with block l .

The blocks accelerate to the right and are constrained to move along the horizontal
surface, so we' ll choose that to be the direction of the x axis. Then Newton's second
law ~ Fy = may for block l tells us that N1 - m 1g = 0, whereas applying~ Fx =
max to block l gives
F - P = m 1ax.
176 FORCES

In this equation there are two unknowns, P and ax, so we need one more equation.
Applying the second law to block 2 (and using rthe constraint that the acceleration has
the same value for both blocks) we get N - m2 g = 0 and
p = m2ax.
Substituting this value for P into our previous equation, we have

which allows us to solve for ax:

ax = F l(m 1 + m 2).
(We could also have obtained this result by thinking of the two blocks as one block of
mass m1 + m 2 acted upon only by the force Fin the x direction.) Substituting numbers,
we find
ax = 0 .67 m/s2
To find the force P that block l exerts on 2, we simply substitute the value of ax
into the equation for P ,
P = ~ax = (2.0 kg)(0.67 m/s 2 ) = 1.33 N.
The direction of P is shown in the free-body diagram.

Example 4
If the mechanism for transmitting push P from block 1 to block 2 in Example 3 is a
massless spring rather than direct contact between the blocks, what is the compression
of the spring? (Assume constant compression, and use spring constant k = 10 3 N/m.)

Let us focus our attention on the forces and motion in the x direction. Since the
spring is massless, the acceleration of the whole system is still ax = F l (m 1 + m2) as in
Example 3. And because the separation between blocks is assumed constant, each block
still accelerates at this same rate.
The forces on block 1, the spring, and block 2 are as indicated in the free-body
diagram:
8.4 APPLICATION OF NEWTON'S LAWS

Applying Newton's law L Fx = max to the spring we obtain

P1 - P 2 = m(spring)ax

But because the mass of the spring is assumed negligible, we have P1 = P2 Newton's
law for each block then takes the same form as in Example 3. We find from Example 3

2 mF
Pz = m2ax = - -=--
ml + m2
for the force on the spring. The compression of the spring is

lxl = p2 = m2F
k (m 1 + m2 )k

Substituting numbers, we find lxl 2115 em.

Example 5
A 3.0-kg monkey holds on to a light rope which passes over a frictionless pulley and is
attached to a 4.0-kg bunch of bananas. What is the acceleration of the monkey?

Since we have two objects, we draw a free-body diagram for each . Because tensions
in ropes always pull, me tension T acting on the monkey is vertically upward. The same
tension T pulls vertically upward on the bananas as well.

Let' s label the monkey's mass, position, and acceleration by m 1 , z., and a 1 measured
vertically upward, and the bananas' mass, position, and acceleration by m2 , z2 , and a 2
measured vertically upward.
Applying L F,, = m 1a 1 to the monkey, we have
(a)

In addition, the fixed length of the rope imposes the constraint that the bananas accelerate
downward at the same rate as the monkey accelerates upward, a2 = - a 1 Using this
result to eliminate a 2 from the second application of Newton's law, we obtain
178 FORCES

(b)

We are left with two equations (a) and (b) and two unknowns. Subtracting (b) from (a)
to eliminate T we find -m 1g + m2 g = m 1a 1 + ~a 1 ; solving for the acceleration a 1
we obtain
a 1 = (m2 - m 1 )gl(m2 + m 1).
Since the bananas are heavier than the monkey (~ > m 1) , the monkey accelerates
upward, while the bananas accelerate downward at the same rate. Substituting values,
we find a 1 = 1.4 m/s 2 The acceleration is substantially less than g because the combined
monkey- rope- banana system behaves like a linear chain, with the pulls at either end
partially canceling whereas the inertia of the monkey combines with that of the bananas.
If the monkey and bananas are replaced by simple weights, this arrangement is called
an Atwood's machine. If m 1 and m 2 are nearly the same, the acceleration can be made
quite small, permitting an accurate measurement which determines g rather well.

Example 6
A puck of mass m on a frictionless table is attached to a mass M by a light string which
passes through a hole in the table. What must be the speed of the puck if it is to move
in a circle of radius r while M remains at rest?

We know that the circular motion of the puck must be sustained by a centripetal
force. Here the tension in the string (created by the weight Mg) provides that force. The
free-body diagrams are shown below:

z
- a=v 2 /r
N
,_j
side
f-r T

~
view

mg

Mg

Since the puck is moving in a circle, it has a centripetal acceleration a = v 21r. That's
how the speed of the puck enters the problem.
ForM let's choose the positive z axis to be vertically upward, and for the puck, we
choose the radial direction outward to be the positive r direction (the direction opposite
8.5 FRICTION 179

to the centripetal acceleration). Applying the second law~ F: = 0 to the mass M (which
is not accelerating), we find
T- Mg = 0.
For the puck, recalling that both the tension and acceleration act inward, ~ F, = ma,
implies - T = - mv 2/r , so
T = mv 2/r.
Substituting for T and solving for the speed, we find
v = VMgrlm.

8.5 FRICTION

In 1699 the French scientist Guillaume Amontons investigated the losses caused by friction
in machines. From his studies he found the empirical relationship that frictional forces
from a surface do not exceed an amount proportional to the normal force exerted by the
surface on the object,
f :5 v.,N,
where v. is the coefficient of friction. Later Coulomb noted that v. varies with the two
materials that are in contact.
Friction is an inescapable example of a force produced by electricity. At times we
wish that we could do away with it, so as, for example, to improve engine performance,
yet without it (a condition approached on ice or a waxed floor) we couldn't walk. Even
though a highly polished object may appear smooth, when examined through a microscope
it appears very rough, having countless surface irregularities, as shown in Fig. . When
two objects are placed in contact, the many contact points resulting from the (microscopic)
rough edges tend to interlock or even become welded together by electric forces. When
one object moves across another, these tiny welds continually rupture and reform. In
addition the interlocking obstacles must be overcome by lifting or deformation or abrasion.
The net result of these complex causes is friction - a force parallel to the surface which
opposes the motion of the object. Since the number of welds is proportional to pressure
from the object on the surface, the force of friction is proportional to the normal force
on the object, a.s Amontons found.

Figure 8.6 Micr<Oscopic examination of a highly polished surface

revea.ls irregularities.
180 FORCES

The frictional forces acting between surfaces at rest with respect to each other are
called forces of static friction. Suppose that you have a block at rest on a horizontal
surface. By Newton's second law, the force of friction is zero, as Fig. illustrates .

(a)
D
RF
(b)

!,

r1 F
(c)

I
!,

(d)
I F = p,N slipping occurs

Figure 8.7 The force of static friction increases up to a maximum

value equal to fJ., N .

Now suppose you apply a small measurable force F to it as in Fig. , and observe
that the block doesn't move. By Newton's second law, the force of static friction is equal
in magnitude and opposite in direction to F. Now suppose that you increase F and note
that the block still doesn't move. The force of static friction increases as well , always
being equal to F in magnitude, as Fig. shows. As F is further increased, there will
be a definite value ofF for which the block slips, as Fig. illustrates . The smallest
force necessary to start motion is the maximum force of static friction. These observations
can be summarized by the relation for the magnitude of force of static friction fs.

Static friction:
f s !5 fl.sN (8.7)

where 1-1-s is the coefficient of static friction which depends on the two surfaces in contact
and N is the normal force. Static friction always acts parallel to the surface and opposes
the intended motion of an object in its rest frame. Table lists values of 1-1-s for various
materials.
Once the block begins to move, kinetic friction acts on the block. This frictional
force is usually less than static friction (e.g., it is harder to form welds on a moving
contact). The magnitude of the kinetic friction A approximately obeys the empirical
relationship
8 .5 FRICTION 181

Kinetic friction:
(8.8)

where N is the normal force and I-Lk is the coefficient of kinetic friction which depends
on the two surfaces in contact. Table lists a few values of I-Lk. The force of kinetic
friction is always opposite to the velocity of the object.

Table 8.2 Coefficients of Static and Kinetic Friction

Material J.l.s ILk

Steel on steel 0.78 0.42

Nickel on nickel 1.10 0.53

Teflon on Teflon 0.04 0.04

Oak on oak (parallel to grain) 0.62 0.48

Oak on oak (perpendicular) 0.54 0.32

Ice on ice 0.05 0.04

Example 7
A spring with pointer attached is used as a force scale. First the spring is calibrated by
hanging a known mass M = I kg from it vertically; the mass stretches the spring (moves
the pointer) 5 em. Then the spring is used to pull a wooden block of mass m = 0.5 kg
horizontally across sandpaper. The spring is stretclhed by 3 em just before the block starts
to move, but only 2 em when it moves at constant velocity. What is the spring constant
k, and what are the coefficients of friction IJ.s and I-Lk?

s
10
15

The equilibrium condition for the calibration is klxl = Mg, where lxl is the distance
the spring is stretched, so the spring constant is
2
k = Mg = (I kg)(9.8 m/s ) =
196-.
N
lxl 0.05 m m
182 FORCES

For the block on the sandpaper, the equations of equilibrium are

N = mg, klxl =f.

mg

Just at the point of slipping, the static friction becomes Is = JJ.sN = JJ.smg. Therefore
= klxl = (196 N/m)(0.03 m) = 12
JJ.s mg (0.5 kg)(9.8 m/s 2 )

When the block moves at constant velocity, the spring force is balanced by kinetic friction,
A= !J.kN = JJ.kmg. Since the spring displacement is now only two-thirds as great, the
force is only two-thirds as great and we have

IJ.k = klxl = 0.8.

mg

Example 8
A block rests on an inclined plane which has a variable angle e. The angle e is increased
from zero, and at 40 the block slips. What is the coefficient of static friction?

L In a free-body diagram for the block, static friction is directed up the plane (because
the block tends to slip downward).

) N~
X mg

Choosing the positive x axis ahng the plane and the positive y axis perpendicular to the
plane, we have the following:
8.5 FRICTION 183

L Fx == 0 implies mg sin 0 - fs = 0 ,
L Fy == 0 implies N - mg cos e == 0.
When 0 == 40, the force of static friction is maximum and we can substitutef5 = 1J.5 N.
But from our second equation, the normal force N == mg cos 40. Substituting all this
into our first equation, we get
mg sin 40 - !J. 5 mg cos 40 = 0 ,
which tells us that
IJ.s = tan 40 = 0.84.
Note that only when maximum static friction is acting on an object can you use fs
IJ.sN.
When 0 is greater than 40, the block slides down the plane, with~ Fx = mg sin 0
!J.kmg cos 6 = ma, giving for the acceleration
a = g(sin e- IJ.k cos 0).
This equation is also sometimes relevant at 0 < 40, but here it must be treated with
care. At small 0 (tan 0 < iJ.k ) it seems to predict an acceleration up the plane. This
correctly describes the case in which an initially downward sliding block decelerates until
it comes to rest, but at that point we must switch over to the equilibrium (a = 0) condition.
Remember that friction opposes motion; it cannot of itself produce motion up the plane.
In the usual situation IJ.s > iJ.k there is also an intermediate range of angles (IJ.s > tan 0
> !J.k) where static equilibrium can be satisfied, but a > 0 once the block is set in motion.
Here equilibrium is only metastable; a sufficiently strong blow will start the block ac-
celerating downwards.

equilibrium metastable sliding

equilibrium

Example 9
A ladder of weight wand length L leans against a wall. The wall is frictionless, but the
ground has coefficient of friction IJ.s At what angle 0 does the ladder start to slip?
184 FORCES

If we assume the ladder is uniform, the weight acts through its center as shown in
the free-body diagram. The force exerted on the ladder by the waH is normal to the wall,
but the force exerted by the ground has components both normal and parallel to the
ground owing to the presence of friction .

Newton's second law gives

2: Fx = 0 implies N1 - f s = 0,
2: Fy = 0 implies N2 - w = 0.
To determine.fs we must also balance torques. Taking torques around the point of contact
with the ground and following the convention that counterclockwise torques- are positive,
we have

w ~ sin 6 - N 1L cos 6 = 0,

so
w
N1 = fs = -tan 6.
2

At the point of slipping, fs = J.Lsw , so

tan 6 = 2J.Ls

Note that the force exerted by the ground is not generally directed along the ladder,
but must pass through the intersection point of N 1 and w to avoid giving a net torque
about that point.
If the wall is rough, its force on the ladder includes a parallel component .fs 1 There
are now four unknown forces, N 1 , N2 , fs, and .fs1> but still only three independent
equations! This simply means that the solution is not unique*; the ladder can be kept in
equilibrium for a given angle 6 by many possible sets of forces.

*To be sure. more equations can be obtained by setting L T1, = 0 about different points. But this does
not provide independent conditions because, as we proved in Section , having L F1 = 0 and L T 1 = 0
about one point ensures that L T, = 0 about all points.
8.6 DRIVING ON CURVED ROADWAYS 185

Friction is messy, and the rules given above are idealizations. To give an example
of the complications that arise in practice, conside.r the coefficient of friction for a copper
block on a copper plate. The value of ILs given in !handbooks is about 1.6. But if we took
great care to prepare clean smooth surfaces, working in a vacuum to avoid oxidation and
prevent even a thin layer of air from getting between the blocks and the plate, the atoms
on the surface would forget which piece they belonged to and would fuse together. The
coefficient of friction would be huge! The value of IL given in the handbook refers to a
typically oxidized, dirty surface as found in normal use. Furthermore , to tell the whole
truth, IL is not generally independent of velocity and N. It is not fundamental at all. The
use of Eqs. and with constant handbook values of IL simply provides a rough
and ready first estimate, which is often all one needs in practice.

8.6 DRIVING ON CURVED ROADWAYS

The motion of an automobile around a curve is an important example bringing into play
many of the principles in this chapter. Consider an automobile rounding a circular curve
whose center is a distance R to the right. The force acting on the car must supply a net
centripetal acceleration v 2 /R to the right.
lf the cu.rve is unbanked, the centripetal acceleration must be supplied entirely by
friction. Figure shows the car, with the reaction F of the road as well as the force
due to gravity drawn (for simplicity) as acting through the center of gravity. In Fig.
we exhibit the vectorial relation that must exist among F, gravity, and centripetal force.
Figure shows the free-body diagram with F split into normal and frictional com-
ponents.

N
F
mu2JR

J7. ~~

mg
a= v 2 /R

mg
(a) (b) (c)

Figure 8.8 Forces on a vehicle rounding an unbanked curve. In (c),

N and fare net forces not drawn at their points of application; the
individual normal and frictional forces applied at each tire will be
shown in Fig.

Newton's second law gives

N - mg = 0, (8.9)

mv 2
J =R- . (8.10)
186 FORCES

The maximum strength of friction* is f = ~sN ~5 mg , so the car goes off course
(skids) if
v2
R > ~s g (8.11)

At high speed, a small R and a slippery road (low ~5 ) are dangerous.

If the curve is banked, less reliance is placed on friction. The reaction force F
required to keep the car on course is exactly the same as before (Fig. , but now
more of it can be supplied by the normal force. Figure shows the optimally banked
case in which the road surface is perpendicular to the required reaction force F and friction
does not come into play at all. In this case Newton's second law gives

mg

(a) (b)

Figure 8.9 Vehicle rounding a banked curve.

N cos 6 - mg = 0, (8.12)

mv 2
N sin 6 = - - (8.13)
R
Solving for the optimal banking angle, one obtains
v2
tan6 = - (8.14)
gR
Engineers construct banked turns with an average speed in mind. Vehicles making the
turn at greater speeds than the road is planned for require a component of friction to stay
on the road, but less than an unbanked road would require.
In reality the road acts on the car not at the center of gravity, but at the points of
contact with the tires as pictured in Fig. . So torques must be considered, and it is

*Here we use the fact that static friction applies in the radial direction when the car bas no velocity
component in that direction.
In the direction of the car's rolling motion a new category of friction applies: rolling friction. Kinetic
friction does not apply because (as will be discussed in Sec. a rolling tire is at rest just at the instantaneous
point of contact with the road . Otherwise it would be skidding! Rolling friction works by a somewhat different
mechanism, involving peeling off of the tire surface and also continual deformation of both the tire and the
road surface underneath, which bas the effect that the tire is always climbing out of a slight depression it has
made. The use of wheels is advantageous because rolling friction is much less than the ordinary kinetic friction
associated with sliding. The motion of a car is opposed by kinetic friction only when the tires skid.
8.7 MOTION IN A RESISTIVE MEDIUM 187

mg
1---- L - ----t
l
Figure 8.10 Torques on a car making an unbanked tum.

here that the virtues of a low center of gravity and broad wheelbase in preventing overturn
become evident. Newton's second law now gives
(8.15)

(8.16)

It is convenient to consider the torques about the center of gravity. They must satisfy

(8.17)

where h is the height of the center of gravity above the road and L is the length of the
axle. This condition applies, in spite of the car's turning motion, because the car does
not rotate in the plane of the diagram. Solving these three equations, we find
mg mv 2 h
Nt = 2 + RL ' (8.18)

(8.19)

The normal forces on the two sides are not equal. If equilibrium requires negative N 2 ,
the road cannot supply it and the car will overturn. This occurs when
v2 gL
R> 2h (8.20)

so large L and small h are desirable for stability. Vehicles are normally engineered to
skid before they overturn on an unbanked curve. Comparing Eqs. and , we
see that this requires
L
2h > ILs (8.21)

8.7 MOTION IN A RESISTIVE MEDIUM

In the Principia Newton considered the motion of objects in resistive media- for example,
dust particles falling through air or marbles fal ling in water. The resistive force in these
cases is viscosity. It is another contact force, extremely complicated to work out in detail.
188 FORCES

But when an object moves at low velocity through a fluid, such as a gas or liquid , the
viscosity is approximately proportional to the velocity. Mathematically we write this as
F vis = - KTJV. (8.22)
The minus sign indicates this force is always opposite to the velocity of the object. The
proportionality constant K depends on the size and shape of the object, and TJ , the
coefficient of vjscosity , depends on the internal friction between different layers of the
fluid.
In general, calculating K is laborious, but a century ago George Stokes found the
result for a sphere of radius R (known as Stokes 's law):
K = 6-rrR . (8.23)
Thus the viscous force acting on a sphere takes the simpler form*
F = - 6-rrRTJV. (8.24)
From Stokes's law we see that K has units of meters, so that by Eq. the
coefficient of viscosity TJ has units of newton-seconds per square meter. This coefficient
depends on temperature: for liquids it decreases with increasing temperature, but for gases
it increases with temperature. Table lists the coefficient of viscosity for several fluids .
A simple laboratory demonstration of the viscous drag force is provided by dropping
a marble into a beaker of a very viscous liquid, like glycerin. The marble appears to fall
with a constant speed. The forces acting on the marble are gravity, mg downward, and
viscous force - 6-rrR'T)v upward. If the marble is falling with a constant speed vL, its
acceleration is zero, so by Newton' s second law,
mg - 6-rrRTJVL = ma = 0. (8.25)

Table 8.3 Coefficients of Viscosity at 20C (Unless Noted)

Liquid "11 (10- 3 N slm 2) Gas "11 (lo- s N slm2 )

Water (0C) 1.792 Air (0C) 1.71

Water 1.005 Air 1.81

Castor oil 9.86 Ammonia 0.97

Glycerine 833 Hydrogen 0.93

*This force law is fairly accurate for the small, relatively slow raindrops and oil drops treated in this
chapter. It is not accur.tte for cannonballs or skydivers. The resistive force for such large objects at their normal
speeds is approximately proportional to their cross-sectional area times the square of the speed. For more
detailed discussion of this subject see, for example, D. Roller and R. Blum, Physics: Volume/, Mechanics.
Waves, and Thermodynamics (Holden-Day, San Francisco, 198 1), pp. 395-404.
:S.7 MOTION IN A RESISTIVE MEDIUM 189

In other words, the viscous force is equal to the weight. Solving for the speed vL, called
the Limiting or terminal velocity, we find
mg
VL = 67TR'T] . (8.26)

Example 10
Find the terminal velocity of a small raindrop assumed to have a radius of 40 IJ.ffi, in
still air.
According to Eq . we need to know 'TJ, R , and the mass of the raindrop. The
'TJ required is that of the medium the drop falls through - air, not water; from Table
we have 'TJ = 1.8 X w-s N s/m2 for air. The radius R = 4 X lO - s misgiven . The
mass is simply the density of water, p = 103 kg/m 3 , times the volume, which is assumed
to be spherical:

Substituting, we have

inserting values, we find v = 0 .2 m/s.

Note that this result is strongly dependent 0111 radius; the sort of fine droplets found
in a mist fall very slowly but large raindrops fall much more rapidly. However, for
raindrops larger than 40 IJ.m, Stokes's law in not accurate because turbulence (not con-
sidered here) sets in. Furthermore, drops of radius >500 1-1-m become nonspherical in
falling. Thus for large drops the detailed relation between vL and R differs from that
given above.

Equation tells us that terminal velocity is proportional to the weight of an

object. In other words, that heavier bodies fall faster! Can this be true? Is the world really
Aristotelian? The answer is that we are including air resistance- precisely what Aristotle
thought must always be present, and Galileo preferred to ignore. But now we have
Newton's laws, which provide the framework to treat the question systematically and to
appreciate the circumstances in which Aristotle's ideas apply and those in which Galileo's
approximation to the full dynamics is accurate.
For any spherical object falling at low' velocity in a viscous medium, not necessarily
at terminal velocity, Newton's second law implies
dv
m dt = mg - 61rR'T]v. (8.27)

This is a differential equation. Presently we will solve it and see how the velocity depends
on time. But first, let's see what we can learn from the differential equation without
solving it.
190 FORCES

Let's look at what Eq. describes. At the instant we drop the object, it is at
rest, which means v = 0; therefore the viscous force - 61rR11v is also momentarily zero.
Consequently, we have m dvldt = mg at that instant. In other words, the object starts
with acceleration dvldt = g, just as Galileo said it should. Because the ball accelerates,
the speed and viscous force increase. As a result, the right-hand side of Eq. , mg
- 61TR1Jv, becomes smaller than mg, and the acceleration decreases. As the velocity
increases (of course, the velocity increases ever more slowly as time goes on because
the acceleration is getting smaller) the right-hand side approaches zero. If it were zero,
gravity and the viscous force would balance each other, so the object would have zero
acceleration and would fall with the terminal velocity.
The object starts out being Galilean and ends up being Aristotelian. A key question
is, how long does this take? If it takes hours to reach terminal velocity, we can forget
about Aristotle; the effects of viscosity can be ignored. But if it takes only a fraction of
a second, then the object spends most of its time falling at terminal velocity.
We can figure out whether the velocity v approaches the terminal velocity slowly or
quickly by a dimensional analysis of the differential equation . This means that we
take a look at the units of each term. If we divide Eq. by the mass m we get
dv 61rR11
- = g - - - v.
dt m

Since each term must have the same units, distance/time2 , the factor multiplying v on
the right must have units lltime, so its reciprocal has units of time. Let's denote this
reciprocal by t 0 . Thus, by definition,
m
(8.28)
to = 61TRTJ '
and t0 has units of time. The differential equation now becomes
dv v
-=g--.
dt t0

What "time" does t 0 represent physically? To fmd out we let .t- oo in the differential
equation. The velocity approaches the terminal velocity vL, and d v!dt approaches 0, so
in the limit we get
VL
O=g - - , or
to
ln other words, t 0 is the time it would have taken to reach the terminal velocity vL if the
acceleration were always equal to g. The number t 0 is called the "characteristic time. "
We will now solve the differential equation and see that the value of t0 will tell us whether
the velocity v approaches terminal velocity vL quickly or slowly, depending on whether
t 0 is small or large.
The differential equation may appear more complicated than any we have
seen so far, but actually it is a familiar one in disguise. Since g = vdto we can rewrite
the differential equation as follows:
8. 7 MOTION IN A RESISTIVE MEDIUM 191

dv v - vl.
-=
dt t0
Now let w = v - vl., the difference between the actual velocity v and the terminal
velocity vl.. Then dw/dt = dvldt since vl. is constant, so w satisfies the simpler differential
equation
dw
- = - - w.
dt t0

This is the differential equation for the exponential function which we've seen earlier in
Chapter (Problem 3. 9). Its solution is
w(t) = w(O)e _,,,o
where w(O) = v(O) - vl.. But v(O) = 0 so w(O) - vl.. Replacing w(t) by v(t) - vl.
we see that
v(t) - vl. = - vl.e- "' 0
so the solution is
(8.29)

The characteristic time t 0 appears in the denominator of the exponential term, so it governs
the rate at which this exponential tends to zero. If t 0 is small, the exponential decays
very rapidly and v quickly approaches the terminal velocity vl. . If t0 is large the exponential
decreases more slowly and it takes a longer time for v to approach vi..
We can now systematically predict which motions will follow Galileo's prescription
and which will appear Aristotelian. A heavy ball (large m) falling in air (very small T)),
according to Eq. , takes a very long time to reach terminal velocity; it tends to
behave in the way Galileo described, unless it falls very far. On the other hand, a marble
(let's say with m = 0.01 kg, R = 0.01 m) in glycerin (T) = 0.8 N s/m2) takes a time
t 0 = 0.06 s to approach terminal velocity; in other words it is close to terminal almost
all of the time we are watching it.

Example 11
If vl. = gt 0 and v(t) = vl.(l - e - 1110), check that v(t) has the physically expected limiting
behaviors: (a) v(t) --+ vl. as t--+ oo and (b) v(t) = gt if tlt0 is small. Find v(t0 ).
(a) As t --+ oo the exponential term goes to 0 and v(t) --+ vl. as expected.
(b) To find a linear approximation for 1 - e-"10 we use the second fundamental
theorem of calculus,

e" - 1 = J: ~ dx.

If u is small the integrand is nearly I so e" - 1 = u, or

I - e" = - u.
192 FORCES

Taking u = - t!t0 we get

v(t) = vl.(l - e - '1' 0) = vLtlt0 = gt
if t!t0 is small. In other words, the object essentialJy undergoes free falJ until the velocity
becomes large enough to make air resistance appreciable. The curve below is the graph
of v(t) and the dashed lines indicate the limiting behaviors.

11(1) free-fall velocity gt

"-/
// terminal velocity
----- ~ ------ --------
///

///
/.

As for v(t0 ), we find by substituting t t0 in Eq.

v(to) = vL(l - e - 1
).

Evaluating the exponential, we find v(t0 ) = 0.63vL. Thus we have another quantitative
interpretation of the time t 0 : it is the time required for the object to reach 63% of its
terminal velocity.

8.8 THE OIL-DROP EXPERIMENT

To conclude this chapter, aJI of its manifold threads- fundamental forces , their precise
strength, contact forces, and the application of Newton's laws of motion- are now going
to be drawn together in recounting one of the most famous experiments of modem physics.
The fundamental electric force on a charge q 1 due to a second charge q 2 ,

F = K q 1q 2 r (8.2)
r2
can be broken into the product of q 1 and an electric field (KA 21rl)r due to the second
charge. The total electric force on q 1 is likewise the product of q 1 and the total electric
field due to the summed effects of alJ other charges. Because the force has this form, a
study of the motion of a charged particle in an electric field can determine its charge.
The electron was discovered by J. J. Thomson in 1896. Nevertheless, because of the
difficulty of observing the motion of a single electron accurately, some time passed before
its charge was determined. Instead of observing a single electron, a more feasible approach
was to study the motion of small droplets that carry only a small net excess of electronic
charges. In 1906 Robert A. Millikan , then an assistant professor at the University of
Chicago, devised an ingenious experiment which for the first time made it possible to
measure the charge on an individual droplet. Through Millikan's experiment it became
possible to verify that electricity in gases and chemical solutions is built out of d iscrete
units of electric charge, and to detennine what that unit is.
8.8 THE OIL-DROP EXPERIMENT 193

Figure 8.11 Robert A. Millikan's original apparatus to measure the

electron charge. (Courtesy of the Archives, California Institute of
Technology.)

Millikan's original apparatus is shown in Fig. . Millikan used oil for the very
same reason mankind spent three hundred years improving clock oils: oil droplets scarcely
evaporate. Therefore the viscous force would not change during an experiment. Sprayed
from an atomizer, the droplets would acquire a charge q due to friction (remember,
friction is a result of e lectrical forces between atoms) as they pass through the nozzle of
the atomizer. The charged droplets then fall through a hole in the uppermost of two metal
plates, which Millikan connected to a lO,OOO-volt battery. While between the plates, the
droplets experience an electric force in addition to gravity, as shown in Fig. . The
electric field E is downward but because of the negative charge of the droplets (and
electrons), the force qE is upward .

.'

c::==J::::(J
telescope
Battery
Figure 8.12 Schematic of Millikan' s oil-drop apparatus.
194 FORCES

Through a viewing device placed a couple of feet from the chamber, Millikan could
watch individual droplets illuminated by light which passed through water so as not to
heat the air in the chamber. Because of their extremely small size, the droplets appear
as stars on a black background. By increasing the voltage, he could make droplets rise;
those droplets with greater charge rose more quickly. By reversing the voltage, he could
make them faJI faster. In addition, he could change the charge on a droplet by sending
a stream of ions into the chamber. Millikan's fascination with the acrobatic motion of
droplets lightened the long, solitary hours he spent iin the lab squinting through the eyepiece
and recording hundreds of measurements.
By adjusting the voltage on the plates (and hence the electric field) certain droplets
could be suspended when the upward electric for<:e equaled the weight of the droplet:
qE = mg. (8.30)

One might think that Millikan could determine the charge q directly from Eq.
But the mass of the droplet also appears in Eq . , and although Millikan knew the
relation

(8.31)

and the density p, the radius R varies from drop to drop. If a drop is small enough to be
balanced by a not overly large field, its radius is too small to measure directly by light.
So Millikan had to determine the radius by a second measurement on the same droplet,
using the rate of fall when the field was reduced. Or more generally, since in practice
the perfect balance of Eq . is hard to achieve, he measured the rate of rise and fall
of a given drop at two different voltages.
Precisely what did Millikan do to determine the charge of the electron? First, he
studied the motion of a droplet drifting upward between the plates of Fig. . According
to Newton's second law this motion can be described by
dv
m- = qE - mg- 6'!TRT]v. (8.32)
dt
In this case the electric force pushes the negatively charged droplets upward but the
viscous force acts downward (opposite v) in the same direction as gravity. By setting
dvldt ;; 0 in Eq . , we find the terminal velocity to be
qE - mg
(8.33)
Vt = 6'1TR1] .

The characteristic time to reach this terminal velocity turns out to be the same as when
there is no electric field and is given by Eq.
m
(8.28)
to = 6'!TRT].

One finds that a typical droplet of size such that qE and mg nearly balance reaches terminal
velocity very quickly, and that the terminal velocity is quite slow. Using a stopwatch to
time a droplet moving between fiducial marks engraved on the viewing device, Millikan
could measure the terminal velocity. When he rewrote Eq . in the form
8.8 THE OIL-DROP EXPERIMENT 195

qE - j1rR 3 pg
(8.34)
61rR11
by using Eq. to relate the mass of the droplet to its radius, only q and R remained
unknown.
A second measurement was provided by turning the electric field off and watching
the free fall of the same droplet. When the droplet is simply falling under the force of
gravity, the terminal velocity is given by Eq.

(8.26)

which we again rewrite as

2R2pg
v2 =-- (8.35)
91]
with the aid of Eq. . In effect, this second measurement is used to find the radius
of the droplet, R.
Using Eq. we can eliminate R in Eq. , and with some algebra solve for
q. The result is
181TTJ312 112
q EViiP v2 (v. + v2). (8.36)

By measuring v 1 and v2 Millikan determined the charge of a droplet.

To be very precise, Millikan actually used p - CT in place of p, where CT is the
density of air. The reason is that air provides an additional upward buoyant force on a
droplet, which is equal to the weight of the air displaced by the droplet (this is known
as Archimedes' law). The weight of air displaced is just the density of air times the
volume of the droplet. Accounting for this force is equivalent to saying that the weight
of the droplet is reduced to mg - m,.g = (m - ma)g, where ma is the mass of air
d isplaced: m3 = j1rR 3CT. Since the density of air is about one-thousandth that of oil, the
correction is barely necessary. With this correction taken into consideration, Eq.
becomes
181TTJ3/2
v~ (vi + v2).
12
q = EV2g(p _ CT) (8.36)

A correction to Stokes's law which we shall not discuss here was also necessary to obtain
precise results.
Through hundreds of delicate measurements, Millikan, the patient experimentalist ,
discovered that the charge on a droplet always comes out an integral multiple (like 1, 2,
3, etc.) of the smaJlest charge he found. Here was confirmation that charges come in
integral multiples of a fundamental charge - the ,c harge of the electron.
By reevaluating the coefficient of viscosity for air, and reducing errors caused by
temperature and air currents, Millikan succeeded in determining the charge e of the
electron with an error of 0. 1%. The value he published in 19 13 was e = - ( 1.603
0.002) x 10 - 9 coulomb, which served physics for a generation and is within ex peri-
196 FORCES

mental error of the most recent value. For his momentous efforts, Millikan received the
Nobel Prize in 1923.
The quest for the fundamental constituents of matter goes on. Today some physicists
are searching for fractionally charged particles called quarks. Based upon a symmetry
classification for elementary particles, quarks are the building blocks of particles which
exist inside nuclei and carry charges of + ~e and -~e. Modifications of Millikan's historic
experiment are used by some of these quark hunters.

8.9 A FINAL WORD

When Millikan made his measurements, alone i.n his laboratory, like any scientist, he
recorded what he had done in a notebook. Afterward, he would gather his results, write
a scientific paper, and publish it for all the world to see. But his notebooks, the raw data
of his experiments, were for his own eyes only. Figure shows a page from Millikan's
notebook. Before we criticize what we see, let's remember what Millikan was doing. He
was measuring, for the first time ever, one of the fundamental constants of nature. His
task was to make his measurements in the most careful , dispassionate way possible, then
publish all of his results so that other scientists could judge whether he'd done it properly.
The page in Figure is dated March 15, 19 I 2. He writes down the temperature and
barometric pressure, then he starts recording data: the times for a droplet to move between
fiducial marks under pure gravity (G), and with the field (F). Then he calculates the
velocities, uses logarithms to multiply them together (he didn't have a hand calculator),
and finally he gets his result.

Figure 8.13 Page from Roben A. Millikan's research notebook.

(Counesy of the Archives, California Institute of Technology.)
PROBLEMS 197

On one page he writes: "One of the best ever ... almost exactly right" - What's
going on here? How can it be right if he's supposed to be measuring something he doesn' t
know? On another page he writes: " Beauty. Publish! " One might expect him to publish
everything! On another page, the usual stuff, then: "4% too low - something wrong."
Not 4% too low but publish anyway, like a good scientist. Then something very revealing:
" ... distance wrong." He's found an excuse for not publishing it. More pages ...
" Beauty, one of the best, " and so on for pages and pages.
Now, you shouldn't conclude that Robert Millikan was a bad scientist. He wasn 't-
he was a great scientist, one of the best. What we see instead is something about how
real science is done in the real world. What Millikan was doing was not cheating. He
was applying scientific judgment. He had a pretty clear idea of what the result ought to
be- scientists almost always think they do when they set out to measure something. So,
when he got a result he didn't like, he wouldn't just ignore it - that would be cheating.
Instead, he would examine the experiments to see what went wrong. Now that seems
reasonable, but it's actually a powerful bias to get the result he wants, because you can
be sure that when he got one he liked, he didn't search as hard to see what went right.
Experiments must be done that way. Without that kind of judgment, the journals
would be full of mistakes. So, then, what protects us from being misled by somebody
whose "judgment" leads to a wrong result? Mainly , it's the fact that sooner or later,
someone else with a different prejudice will make another measurement. You see there
is a real answer; it's part of nature. And, so long as that's true, sooner or later the truth
will come out. Much is written in textbooks about the scientific method, especially tnat
picking the results you like is a cardinal sin. Don' t believe everything you read. Science
is a difficult and subtle business, and there is no method that assures success .

Problems

The Strength of Gravitational and Electric Forces

1. Using the values of G, g, and the radius of the earth, calculate the mass of the
earth . From your value for the mass determine the density of the earth by treating
the earth as a solid sphere. The average density of rocks on the earth's surface is
about 2. 7 g/cm 3 ; comparing this value to your calculated value, what can you con-
clude about the interior of the earth?
2. Compare the electric force between the proton and electron in a hydrogen atom
with the gravitational force between them (mp = 1.67 X 10- 27 kg, me = 9.11
X 10- JI kg).

Contact Forces

3. A book rests on a table which stands on the floor.

(a) Is the normal force acting on the book the reaction force to its weight? Explain .
(b) List the vertical forces acting on the table. How strong is the normal force
exerted. by the floor on the table?